Flynn et al. Arthritis Research & Therapy 2013, 15:R220 http://arthritis-research.com/content/15/6/R220

RESEARCH ARTICLE Open Access Association analysis of the SLC22A11 (organic anion transporter 4) and SLC22A12 (urate transporter 1) urate transporter locus with gout in New Zealand case-control sample sets reveals multiple ancestral-specific effects Tanya J Flynn1, Amanda Phipps-Green1, Jade E Hollis-Moffatt1, Marilyn E Merriman1, Ruth Topless1, Grant Montgomery2, Brett Chapman2, Lisa K Stamp3, Nicola Dalbeth4 and Tony R Merriman1*

Abstract Introduction: There is inconsistent association between urate transporters SLC22A11 (organic anion transporter 4 (OAT4)) and SLC22A12 (urate transporter 1 (URAT1)) and risk of gout. New Zealand (NZ) Māori and Pacific Island people have higher serum urate and more severe gout than European people. The aim of this study was to test genetic variation across the SLC22A11/SLC22A12 locus for association with risk of gout in NZ sample sets. Methods: A total of 12 single nucleotide polymorphism (SNP) variants in four haplotype blocks were genotyped using TaqMan® and Sequenom MassArray in 1003 gout cases and 1156 controls. All cases had gout according to the 1977 American Rheumatism Association criteria. Association analysis of single markers and haplotypes was performed using PLINK and Stata. Results: A haplotype block 1 SNP (rs17299124)(upstreamofSLC22A11) was associated with gout in less admixed Polynesian sample sets, but not European Caucasian (odds ratio; OR = 3.38, P =6.1×10-4; OR = 0.91, P =0.40, respectively) sample sets. A protective block 1 haplotype caused the rs17299124 association (OR = 0.28, P =6.0×10-4). Within haplotype block 2 (SLC22A11) we could not replicate previous reports of association of rs2078267 with gout in European Caucasian (OR = 0.98, P = 0.82) sample sets, however this SNP was associated with gout in Polynesian (OR = 1.51, P = 0.022) sample sets. Within haplotype block 3 (including SLC22A12) analysis of haplotypes revealed a haplotype with trans-ancestral protective effects (OR = 0.80, P = 0.004), and a second haplotype conferring protection in less admixed Polynesian sample sets (OR = 0.63, P = 0.028) but risk in European Caucasian samples (OR = 1.33, P =0.039). Conclusions: Our analysis provides evidence for multiple ancestral-specific effects across the SLC22A11/SLC22A12 locus that presumably influence the activity of OAT4 and URAT1 and risk of gout. Further fine mapping of the association signal is needed using trans-ancestral re-sequence data.

* Correspondence: [email protected] 1Department of Biochemistry, University of Otago, 364 Leith St, North Dunedin 9016, New Zealand Full list of author information is available at the end of the article

© 2013 Flynn et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Flynn et al. Arthritis Research & Therapy 2013, 15:R220 Page 2 of 11 http://arthritis-research.com/content/15/6/R220

Introduction (OR = 1.02, P = 0.80; in very strong LD with rs2078267 Hyperuricemia is a primary risk factor for gout [1], leading (r2 = 0.96)) with gout. To our knowledge, no previous to the formation of monosodium urate crystals and gout in studies have tested for association of SLC22A11 with SU some people. Genome-wide association studies (GWAS) and gout in non-European populations. have identified 28 loci that account for approximately 6% There is ambiguity regarding association of SLC22A11 of the variance in serum urate (SU) levels in European and SLC22A12 with gout - previous reports have studied Caucasians [2]. Loci with the strongest effect encode pro- different population groups, with various methods for de- teins involved in secretion and renal filtration of . termination of gout; clinical ascertainment using 1977 Two of these, SLC2A9 and ABCG2, have a very strong American Rheumatism Association (ARA) criteria [15,17], effect in gout (risk allele odds ratio (OR) >2.0) in multiple self-report of physician-diagnosed gout, use of gout medi- ancestral groups [2-5], whilst a third locus SLC17A1, has a cations [2,18], a combination of self-report and exami- weaker effect (OR <1.5) [2,6,7]. The SLC22A11 (OAT4) nation of medical records [16]. Here our aim was to and SLC22A12 (URAT1) are also associated with SU clarify the association with gout according to the ARA at a genome-wide level of significance [2]. However, evi- classification criteria at the SLC22A11-SLC22A12 locus in dence for association with gout is ambiguous. multiple ancestral groups drawn from the New Zealand SLC22A11 (OAT4) and SLC22A12 (URAT1) encode (NZ) population, including Māori and Pacific Island. renal urate transporters located together on These groups have a prevalence of gout double that of 11. SLC22A12 is expressed across multiple tissue types European Caucasians [21], with earlier onset and more and developmental stages, whilst SLC22A11 expression is severe gout and a higher prevalence of co-morbidities [6]. restricted to the kidneys and placenta [8]. URAT1 is a high-affinity apical urate transporter [9], whilst OAT4 transports multiple organic anions and has low affinity for Methods urate [10]. Both URAT1 and OAT4 may play a role in Study participants gout secondary to diuretic use due to competitive trans- All gout cases (Additional file 1: Table S1; n = 1,003) had a port of uric acid and diuretics [9,10]. confirmed diagnosis of gout as defined by the 1977 ARA Multiple single nucleotide polymorphisms (SNPs) at the preliminary classification criteria for acute gout [22] and SLC22A11/SLC22A12 locus have been associated with SU were recruited from community-based settings, primary in European Caucasians, with two independent effects re- and secondary care. Controls (n = 1,156) self-reported ported within the locus [2]. These SNPs are spread across having no history of arthritis and were convenience-sam- a 1.3-Mb region and constitute four distinct linkage dis- pled from workplaces and community focal points in the equilibrium (LD) blocks, two of which contain SLC22A11 Auckland region of NZ. Co-morbid conditions (type 2 dia- and SLC22A12 [2]. This suggests a complex pattern of betes, renal disease, hypertension, dyslipidemia and heart association with SU in European Caucasians, at least. In disease) were self-reported. Ethical approval was given by non-European populations, studies specifically on the the NZ Multi-Region Ethics Committee and all partici- SLC22A12 have revealed associations with SU in pants provided written informed consent for the collec- Chinese, Korean and Japanese sample sets that are con- tion of samples and subsequent analysis. sistent with the those reported in European Caucasians Participants were divided into four sample sets ac- (rs505802 and SNPs in LD) [11-14] whilst additional asso- cording to self-reported grandparental ancestry; NZ ciations are evident in Chinese (rs475688 and rs7932775, European Caucasian (420 cases, 638 controls), Eastern in low LD with rs505802) [14,15]. Polynesian (EP) - NZ Māori and Cook Island Māori Variants in the haplotype block that includes SLC22A12 (315 cases, 349 controls), Western Polynesian (WP) - have previously been tested for association with gout Tongan, Samoan and Niuean (249 cases, 144 controls), in European Caucasian with no association of rs505802 and a small mixed eastern and western ancestry group (OR = 0.99) [16] or rs478607 (OR = 0.97) [2]. There is con- (EP/WP) (19 cases, 25 controls) (Additional file 1: flicting evidence for association of common SLC22A12 Table S1). These groupings were based on previous variants with gout in Chinese sample sets [15,17]. Rare evidence for genetic heterogeneity between Eastern variants in the SLC22A12 locus have been associated and Western Polynesia (see [5] and references therein). with SU and gout in African-Americans [18] and gout TheEPsampleset(predominantlyNZMāori) was fur- in Japanese [19] and Micronesians [20]. At SLC22A11, ther split into two groups according to Polynesian Köttgen et al. [2] reported association of the minor (C) al- ancestry estimates obtained using 67 genomic control lele of rs2078267 with prevalent gout in a meta-analysis of markers [6]. Individuals of higher EP ancestry formed European Caucasian cases nested within population-based the EP(High) sample set (256 cases, 187 controls) and cohorts (OR = 1.16, P =4.4×10-6). However, Stark et al. those of lower ancestry formed the EP(Low) sample set [16] reported no evidence for association of rs17300741 (59 cases, 162 controls). Flynn et al. Arthritis Research & Therapy 2013, 15:R220 Page 3 of 11 http://arthritis-research.com/content/15/6/R220

SNP selection wide significance a Bonferroni correction was applied to Data on 1,000 genomes [23] for Caucasian (CEU) and P-values. A conservative correction factor of 48 was Han Chinese (CHB), as the most closely related and avail- applied to the single SNP analysis (P <0.00104; 12 SNPs able population to the NZ Māori and Pacific populations, meta-analyzed by inverse-variance weighting over four was used to define haplotype blocks based on all the SNPs sample sets - EP(High), WP, EP/WP), all Polynesian (in- previously associated with SU in GWAS studies within cluding EP(Low)), European Caucasian and EP(Low), the 1.3-Mb region of (Additional file 2: and all five sample sets combined) and a factor of 12 for Figure S1). Four distinct haplotype blocks were identified the haplotype analysis (P <0.0042; meta-analysis of three (block 1: approximately 26Kb upstream, Block 2: SLC2 haplotype blocks over four sample sets). Interaction terms 2A11, Block 3: including SLC22A12, and block 4: approxi- for co-morbid phenotypes and ancestry were included in mately 840Kb downstream) (Additional file 1: Table S2) the logistic regression model where appropriate. and this information was then used to select variants for Ancestrally defined sample sets were combined by genotyping (Additional file 2: Figure S2). SNPs were se- an inverse-variance weighted fixed-effect method, using lected largely on the basis of previous literature reports of STATA 8.0 software [27]. A Q-statistic was calculated to association with SU. For block 1 and block 4 both the determine the heterogeneity between cohorts and for SNP most associated with SU (rs17299124; β = −3.75 SNPs showing heterogeneity (PHet <0.05) the fixed-effect μmolL-1, P =8.6×10-16 and rs642803; β = −2.56 μmolL-1, model was replaced with a random-effect model. Power P =4.5×10-14, respectively) and the most associated was calculated for a nominal P = 0.05 (Additional file 2: tag-SNP (in high LD with all other associated SNPs Figure S5) [28]. All sample sets were adequately powered (rs475414; β = −2.38 μmolL-1, P =3.1×10-12 and rs122 (>80%) to detect moderate effects (OR = 1.5) at minor 89836; β = −2.62 μmolL-1, P =6.6×10-14, respectively) allele frequency >0.1. were chosen [2,18,24]. For block 2 (SLC22A11) one tag- SNP common to CEU and CHB was identified (rs693591) Haplotype association analysis and combined with the previously studied rs17300741 and Haplotype imputation by expectation maximization was rs2078267 [2,24]. One tag-SNP for block 3 (SLC22A12) performed in PLINK to produce most-likely haplotype was selected (rs476037) and included with the previously calls (haplotypes with a frequency <0.01 were excluded). reported SNPs rs3825018 (in complete LD with rs505802), These data were then used to perform a sex-adjusted lo- rs475688 and rs7932775 [2,24]. AfifthSNP(rs478607) gistic regression in PLINK, producing ORs and P-values. from block 3 was also added as representative of a second Inverse-variance weighted fixed-effect analysis in STATA (weaker) effect potentially independent of rs2078267 8.0 was performed to combine the independent datasets (SLC22A12) ([2]; β = −2.80 μmolL-1, P =4.4×10-11) (as described for the allelic association analysis). (Additional file 2: Figure S3, S4). Three of the twelve chosen SNPs have a predicted biological function according to Locus-specific gout genetic risk score FuncPred [25]; rs3825018 is situated in a potential tran- A weighted genetic risk score was calculated using the scription factor binding site and may enhance or silence most associated SNP in each of block one to three. The splicing, rs7932775 may also enhance or silence splicing gout risk score calculation was ((rs17299124(C) × β)+ and rs476037 is situated at a potential miRNA binding site. (rs2078267(C) × β)+(rs3825018(G) × β)), where each SNP ID followed by an allele in brackets denotes the Genotyping allele count per individual, and the β after each SNP SNPs were genotyped using either TaqMan® assays (Applied denotes the natural log of the ancestry-specific OR for Biosystems, Foster City, CA, USA) in a Lightcycler® 480 that allele. Individuals who did not have complete geno- Real-Time PCR System (Roche, Indianapolis, IN, USA) type information (9.22%) were excluded. (rs693591 (assay ID C_925815_10), rs17300741 (C_323 The mean genetic risk score was calculated for each 4214_10), rs3825018 (C_27162391_10)), rs475688 (C_9257 sample set and a two-tailed t-test performed to identify any 97_10), rs476037 (C_3108662_20), rs7932775 (C_3108 significant differences between case and control values. A 664_10), or the MassArray system (Sequenom iPLEXassay, further sex-adjusted logistic regression was performed to San Diego, CA, USA): rs17299124, rs642803, rs475414, test association with gout. All risk-score analysis was con- rs12289836, rs478607 and rs2078267. ducted in STATA 8.0.

Association analysis Results Allelic ORs were obtained from a sex-adjusted logistic Allelic associations regression in PLINK [26]. These were relevant to the The European Caucasian sample set revealed nominally allele in-phase with the rs3825018 G allele as determined significant (P <0.05) associations with gout at two block- using 1,000 genomes LD data. To assess experiment- 3 SNPs: rs475688 and rs7932775 (SLC22A12; OR = 1.26, Flynn et al. Arthritis Research & Therapy 2013, 15:R220 Page 4 of 11 http://arthritis-research.com/content/15/6/R220

-6 P = 0.043; OR = 1.32, P = 0.033, respectively) (Table 1). ancestry for block-1 SNP rs17299124 (Pc =9.9×10 )and -5 The EP(High) sample set revealed four nominally signifi- block-2 SNP rs2078267 (Pc =2.6×10 ). The interaction cant associations within block 1 (rs17299124; OR = 2.77, with ancestry at rs17299124 was consistent with the P=0.019), block 2 (SLC22A11) (rs693591; OR = 1.41, PHet = 0.01 in the inverse-variance weighted analysis with P = 0.031), and block 3 (SLC22A12) (rs3825018; OR = 1.46, the rs2078267 association also specific to Polynesian P = 0.028, and rs476037; OR = 1.56, P = 0.022). The EP (Table 2). The PInteraction at rs3825018 was 0.033. (Low) sample set showed no significant associations where- as the WP sample set revealed association with the block-1 Haplotype association analysis SNP rs17299124 (OR = 5.65, P = 0.011) and the block-3 We initially combined by inverse-variance weighting hap- SNP rs7932775 (OR = 1.49, P = 0.019). The mixed Eastern lotypes in the same combinations of sample sets as used in and Western Polynesian sample set revealed a nominally Tables 2 and 3. At block 1 there was a haplotype (rs475414 significant association at rs693591 (OR = 4.55, P=0.025) (allele T) - rs17299124 (allele A)) that exhibited significant (Table 1). heterogeneity (P = 0.007) in the combined analysis of all Within block 1, observing that the effect of rs1729914 Het ancestries. Consistent with the rs17299124 data the effect, was stronger in the less admixed Polynesian sample sets which was protective, was restricted to the less admixed (EP(High), WP, EP/WP) (Table 1), we combined these Polynesian sample sets (OR = 0.28, P =6.0×10-4). This sample sets by inverse-variance weighting for rs17299124 protective T-A haplotype was rarer in the less admixed and tested for association with gout in these sample Polynesian sample set (Table 4 control frequency sets, revealing experiment-wide evidence for association 0.044-0.052) compared with EP(Low) 0.203 and European (Table 2; OR = 3.38, P =6.1×10-4). The heterogeneity Caucasian 0.245. P-value of 0.010 (in the combined analysis of all ances- At block 2 no haplotype association was observed. At tries) suggested an ancestral specific effect. Based on this, block 3 the only other experiment- wide significant asso- all SNPs were combined by inverse-variance weighting in ciation was observed, a protective haplotype in the com- four groups - less admixed Polynesian (EP(High), WP, bined sample set (A-C-T-G-A; OR = 0.80, P =0.004).This EP/WP), all Polynesian (including EP(Low)), European haplotype was more common in European Caucasian Caucasian and EP(Low), and the five sample sets com- (control frequency of 0.717) than Polynesian (control fre- bined. Of the other SNPs, only block-3 rs3825018 ap- quency range of 0.303-0.545). The block-3 G-T-C-G-G proached experiment-wide significance in all ancestries haplotype showed evidence for heterogeneity (P = combined (OR = 1.27,P= 0.002). Because no nominal Het 0.014) in the combined analysis of all ancestries – closer association with block 4 was seen in any sample set the examination showed that this owed to a risk effect relevant SNPs (rs12289836 and rs642803) were not in- in European Caucasian and more admixed Polynesian cluded in further analyses. (EP(Low)) (OR = 1.41, P = 0.007) and a protective effect in Block-1 to −3 SNPs rs17299124, rs2078267 and rs38 less admixed Polynesian (OR = 0.63, P =0.028)(Tables3 25018 were tested for association with SU levels in con- and 4). The frequency of this haplotype varied conside- trols (Additional file 1: Table S3). The only nominally rably, even within the Polynesian sample sets (Table 4 significant associations observed (P <0.05) were with control frequency: WP 0.034, EP(High) 0.139, EP(Low) rs2078267 in less admixed Polynesian and total Polynesian 0.131, European Caucasian 0.133). (β = 0.044 mmolL-1, P = 0.007 and β = 0.030 mmolL-1, P = 0.022 respectively). The minor allele increased SU, consistent with the increased risk mediated for gout by Gout genetic risk-score association the minor allele in these sample sets (Table 2). These asso- A weighted genetic risk score was calculated to deter- ciations were, however, not significant after correcting mine the cumulative effect of the approximately 360-Kb P-values for multiple testing (n = 12). (block 1 to block 3) region on gout risk, using the most Interaction terms were included in the sex-adjusted associated marker in each ancestral block. Genetic logistic regression analyses in order to test for interaction risk scores ranged from 0 to 18.03 with no European between genotype (of the most associated SNP in blocks 1 Caucasian individual risk score being >7 (range 0 to to 3), gout and body mass index, type 2 diabetes, dyslipi- 6.21). In each sample set the average case risk-score was demia, heart disease and hypertension in the European significantly larger than the average control risk-score Caucasian, less admixed Polynesian and total Polynesian (Table 5). There was a significant association between samples (Additional file 1: Table S4). There were no sig- gout and the genetic risk score in EP(High) (OR = 1.22, nificant interactions observed (after correcting for the P = 0.002) and WP (OR = 1.14, P = 0.041) but not in number of tests done, n = 45). We also tested for inter- European Caucasian (OR = 1.01, P = 0.80), EP(Low) action with ancestry with significant interaction observed, (OR = 1.03, P = 0.77) or the small EP/WP sample set after correcting for the number of tests done (n = 3), with (OR = 1.32, P =0.21). http://arthritis-research.com/content/15/6/R220 Table 1 Allelic association analysis of the twelve chromosome 11 SNPs with gout Flynn

East Polynesian (EP) (High) East Polynesian (Low) West Polynesian (WP) East/West Polynesian European Caucasian Therapy & Research Arthritis al. et SNPb Chr:Posc Effect/ Allele Freq. ORe Pe Allele Freq. ORe Pe Allele Freq. ORe Pe Allele Freq. ORe Pe Allele Freq. ORe Pe Otherd (95% CI) (95% CI) (95% CI) (95% CI) (95% CI) Case Con Case Con Case Con Case Con Case Con 1a T/C 0.571 0.531 1.19 0.257 0.566 0.601 1.01 0.957 0.403 0.409 1.06 0.758 0.500 0.400 1.65 0.412 0.566 0.564 1.02 0.868 (0.88,1.61) (0.63,1.64) (0.75,1.49) (0.50,5.48) (0.84,1.24) 1 rs17299124 11:64256859 C/A 0.976 0.950 2.77 0.019 0.821 0.801 1.20 0.577 0.994 0.957 5.65 0.011 0.974 0.947 3.00 0.461 0.733 0.752 0.91 0.401 (1.18,6.50) (0.63,2.32) (1.49,21.39) (0.16,55.71) (0.73,1.14) 2 rs693591 11:64325069 C/G 0.522 0.449 1.41 0.031 0.315 0.291 1.13 0.630 0.373 0.367 1.00 0.992 0.526 0.260 4.55 0.025 0.139 0.135 0.95 0.733 (1.03,1.92) (0.68,1.89) (0.72,1.40) (1.21,17.06) (0.72-1.26) 2013, 2 rs17300741 11:64331462 A/G 0.819 0.789 1.30 0.187 0.604 0.557 1.29 0.312 0.808 0.830 0.76 0.213 0.806 0.750 1.42 0.595 0.476 0.439 1.05 0.611

(0.88,1.91) (0.78,2.14) (0.49,1.17) (0.39,5.24) (0.87,1.27) 15 2 rs2078267 11:64334114 C/T 0.938 0.910 1.78 0.063 0.651 0.606 1.19 0.526 0.980 0.944 2.20 0.071 0.921 0.886 1.21 0.815 0.479 0.456 0.98 0.816 :R220 (0.97,3.26) (0.70,2.00) (0.94,5.17) (0.24,6.11) ([0.81,1.18] 3 rs3825018 11:64358809 G/A 0.739 0.674 1.46 0.028 0.500 0.447 1.37 0.206 0.726 0.712 1.16 0.404 0.842 0.660 1.83 0.316 0.305 0.267 1.22 0.065 (1.04,2.06)] (0.84,2.23) (0.82,1.66) (0.56,5.98) (0.99,1.50) 3 rs475688 11:64364291 T/C 0.433 0.427 1.10 0.543 0.370 0.295 1.66 0.073 0.409 0.461 0.78 0.145 0.556 0.420 1.72 0.391 0.269 0.236 1.26 0.043 (0.80,1.52) (0.95,2.89) (0.56,1.09) (0.50,5.97) ([1.01,1.57) 3 rs7932775 11:64367862 C/T 0.496 0.478 1.03 0.866 0.373 0.320 1.36 0.278 0.489 0.415 1.49 0.019 0.500 0.396 1.00 1.000 0.213 0.174 1.32 0.033 (0.76,1.40) (0.78,2.35) (1.07,2.09) (0.35,2.90) (1.02,1.69) 3 rs476037 11:64369540 A/G 0.254 0.206 1.56 0.022 0.149 0.136 1.10 0.787 0.250 0.303 0.78 0.165 0.368 0.250 2.58 0.161 0.088 0.089 1.03 0.846 (1.07,2.28) (0.56,2.16) (0.55,1.11) (0.69,9.68) (0.74,1.43) 3 rs478607 11:64478063 G/A 0.241 0.268 0.85 0.359 0.279 0.184 1.72 0.080 0.196 0.166 1.48 0.085 0.211 0.225 0.82 0.766 0.167 0.144 1.27 0.077 (0.60,1.20) (0.94,3.15) (0.95,2.31) (0.23,2.98) (0.98,1.66) 4 rs12289836 11:65436888 A/G 0.621 0.622 0.94 0.711 0.660 0.631 1.07 0.810 0.693 0.746 0.80 0.237 0.790 0.725 1.22 0.766 0.645 0.683 0.85 0.109 (0.68,1.30] (0.62,1.85) (0.55,1.16)] (0.34,4.40) (0.69,1.04) 4 rs642803 11:65560620 T/C 0.454 0.458 1.00 0.975 0.500 0.474 1.26 0.364 0.413 0.421 0.92 0.624 0.500 0.368 0.71 0.620 0.448 0.474 0.93 0.467 (0.73,1.36] (0.77,2.07) (0.66,1.29) (0.18,2.75) (0.77,1.13) aLD block as shown in Additional file 2: Figure S1 and S2, block 1: upstream, block 2: SLC22A11, block 3: SLC22A12, block 4: downstream. brs2078267, rs3825018 and rs476037 showed some evidence for a departure from Hardy-Weinberg equilibrium in the WP controls (P = 0.007, 0.001 and 0.006, respectively), rs2078267 in the EP/WP cases (P = 0.005) and rs476037 in the European Caucasian cases (P=0.023). None were significant after correction for multiple testing (P <0.00042 (correction factor of 120; 10 datasets × 12 SNPs)). cBase position is according to Genome Reference Consortium build 37 (GRCh37). dEffect/Other, alleles; all odds ratios (ORs) are calculated relevant to the effect allele defined as the allele in-phase with rs3825018 G allele according to 1,000 Genomes linkage-disequilibrium data. eOR, allelic OR (adjusted for sex); P, allelic asymptotic P-value. Genotyping success rate: rs475414: 97.55%; rs17299124: 94.35%; rs693591: 98.75%; rs17300741: 97.59%; rs2078267: 95.41%; rs3825018: 99.40%; rs475688: 98.75%; rs7932775: 92.40%; rs476037: 99.17%; rs478607: 98.01%; rs12289836: 97.82%; and rs642803: 97.73%. SNP, single nucleotide polymorphism; freq., frequency; Con, control. ae5o 11 of 5 Page Flynn et al. Arthritis Research & Therapy 2013, 15:R220 Page 6 of 11 http://arthritis-research.com/content/15/6/R220

Table 2 Inverse-variance weighted analysis of allelic associations European Caucasian and High ancestry Polynesianf All ancestriesg low ancestry Polynesiand Polynesiane SNP Chr:Posb Effect/ ORh Ph Het Ph ORh Ph Het ORh Ph Het ORh Ph Het Otherc Ph Ph Ph (95% CI) (95% CI) (95% CI) (95% CI) 1a rs475414 11:64241844 T/C 1.02 0.862 0.990 1.15 0.233 0.727 1.12 0.270 0.839 1.07 0.376 0.861 (0.85,1.22) (0.92,1.43) (0.92,1.34) (0.93,1.23) 1 rs17299124 11:64256859 C/A 0.94 0.539 0.425 3.38 6.1 × 10,4 0.674 1.95 0.006 0.153 1.71 0.109 0.010 (0.76,1.16) (1.69,6.79) (1.21,3.15) (0.89,3.28) 2 rs693591 11:64325069 C/G 0.99 0.942 0.559 1.25 0.051 0.053 1.23 0.047 0.113 1.12 0.167 0.087 (0.78,1.26) (1.00,1.57) (1.00,1.51) (0.95,1.33) 2 rs17300741 11:64331462 A/G 1.08 0.402 0.446 1.04 0.793 0.173 1.10 0.469 0.254 1.07 0.396 0.388 (0.90,1.29) (0.78,1.38) (0.86,1.40) (0.92,1.24) 2 rs2078267 11:64334114 C/T 1.00 0.999 0.499 1.84 0.012 0.804 1.51 0.022 0.591 1.08 0.370 0.169 (0.84,1.20) (1.14,2.95) (1.06,2.14) (0.91,1.28) 3 rs3825018 11:64358809 G/A 1.24 0.028 0.660 1.33 0.020 0.568 1.34 0.008 0.767 1.27 0.002 0.822 (1.02,1.50) (1.05,1.69) (1.08,1.66) (1.10,1.48) 3 rs475688 11:64364291 T/C 1.31 0.011 0.360 0.95 0.683 0.216 1.03 0.768 0.096 1.13 0.109 0.093 (1.06,1.60) (0.76,1.20) (0.84,1.27) (0.97,1.32) 3 rs7932775 11:64367862 C/T 1.32 0.017 0.923 1.21 0.095 0.256 1.23 0.051 0.412 1.26 0.004 0.550 (1.05,1.66) (0.97,1.51) (1.00,1.51) (1.08,1.48) 3 rs476037 11:64369540 A/G 1.05 0.770 0.874 1.25 0.480 0.014 1.19 0.461 0.036 1.08 0.423 0.070 (0.78,1.40] (0.67,2.34) (0.75,1.89) (0.89,1.31) 3 rs478607 11:64478063 G/A 1.33 0.020 0.371 1.04 0.785 0.148 1.13 0.337 0.109 1.19 0.056 0.167 (1.05,1.70) (0.79,1.36) (0.88,1.44) (1.00,1.43) 4 rs12289836 11:65436888 A/G 0.87 0.156 0.435 0.89 0.327 0.717 0.91 0.423 0.791 0.88 0.085 0.863 (0.72,1.05) (0.70,1.13) (0.73,1.14) (0.76,1.02) 4 rs642803 11:65560620 T/C 0.97 0.728 0.267 0.95 0.662 0.862 1.00 0.982 0.726 0.96 0.586 0.818 (0.81,1.16) (0.76,1.19) (0.81,1.23) (0.84,1.11) aLD block as shown in Additional file 2: Figure S1 and S2, block 1: upstream, block 2: SLC22A11, block 3: SLC22A12, block 4: downstream. bBase position is according to Genome Reference Consortium human genome build 37 (GRCh37). cEffect/Other, effect allele and other allele, all odds ratios (ORs) are calculated relevant to the effect allele defined as the allele in-phase with rs3825018 G allele according to 1,000 genomes linkage disequilibrium data. dCombining EP(L) and NZ European Caucasian sample sets. eCombining the EP(H), WP and EP/WP sample sets. fCombining the Eastern Polynesian EP high (EP(H)), EP low (EP(L)), Western Polynesian (WP) and EP/WP sample sets. gCombining the EP(H), EP(L), WP, EP/WP and NZ European Caucasian sample sets. hOR: inverse-variance weighted analysis OR (adjusted for sex); P: inverse-variance weighted analysis P-value; Het P: heterogeneity P-value, inverse-variance weighted analyses were reanalyzed using a random-effects model if the Het P was <0.05.

Discussion explanation for why Polynesian people exhibit higher SU We analyzed three haplotype blocks - a region upstream and a higher prevalence of gout [6,21]. of SLC22A11 containing no known genes (block 1), The block-1 region has previously never been studied SLC22A11 (block 2) and SLC22A12 (block 3). Collectively for association with gout. The single SNP data of our data indicate a complex pattern of association that is rs17299124 showed the strongest effect in the less ancestry-dependent, with risk appearing to be driven by admixed Polynesian sample sets (OR = 3.38) with non- protective haplotypes that are at a higher frequency in significant effects closer to OR = 1.00 in the more European Caucasian than in people of Polynesian ances- admixed Eastern Polynesian and European Caucasian try. The genetic risk-score analysis (Table 5) emphasized sample sets (OR = 1.20 and 0.91, respectively). This SNP the relative important of the locus on the risk of gout in is therefore a novel candidate risk variant to test for less admixed Polynesian (% variance explained 1.4 to 1.9) association with gout in South East Asian populations compared to the more admixed Polynesian and European (ancestrally related to Polynesian). Haplotype analysis Caucasian sample sets (<0.1% variance explained). Com- indicated that this effect in less admixed Polynesians is mon variation in this locus is likely to be part of the driven by a genetic variant that protects from gout. The Flynn et al. Arthritis Research & Therapy 2013, 15:R220 Page 7 of 11 http://arthritis-research.com/content/15/6/R220

Table 3 Inverse-variance weighted analysis of block- 1 to -3 haplotypes European Caucasian and low ancestry Polynesianb High ancestry Polynesianc Polynesiand All ancestriese Hapa ORf Pf Het Pf ORf Pf Het ORf Pf Het ORf Pf Het Pf Pf Pf (95% CI) (95% CI) (95% CI) (95% CI) Block 1 (rs475414|rs17299124) C|C 0.94 0.501 0.795 0.89 0.305 0.884 0.91 0.349 0.936 0.92 0.242 0.978 (0.78,1.13) (0.71,1.11) (0.74,1.11) (0.80,1.06) T|A 1.07 0.549 0.379 0.28 6.0 × 10,4 0.512 0.50 0.006 0.120 0.56 0.103 0.007 (0.86,1.32) (0.14,0.58) (0.31,0.82) (0.28,1.12) T|C 1.02 0.848 0.584 1.30 0.022 0.914 1.29 0.019 0.965 1.14 0.093 0.572 (0.83,1.25) (1.04,1.63) (1.04,1.56) (0.98,1.33) Block 2 (rs693591|rs17300741|rs2078267) C|A|C 0.97 0.800 0.726 1.24 0.070 0.128 1.20 0.083 0.219 1.10 0.253 0.179 (0.76,1.24) (0.98,1.56) (0.98,1.49) (0.93,1.30) G|A|C 1.08 0.468 0.842 0.78 0.037 0.446 0.83 0.083 0.368 0.94 0.440 0.204 (0.88,1.31) (0.61,0.99) (0.67,1.03) (0.81,1.10) G|G|C , , , 1.35 0.093 0.369 1.31 0.113 0.509 , , , (0.95,1.91) (0.94,1.83) G|G|T 0.97 0.706 0.682 0.57 0.022 0.797 0.69 0.042 0.600 0.91 0.252 0.316 (0.81,1.15) (0.35,0.92) (0.49,0.99) (0.77,1.07) Block 3 (rs3825018|rs475688|rs7932775|rs476037|rs478607) A|C|T|G|A 0.82 0.040 0.953 0.76 0.035 0.272 0.77 0.024 0.451 0.80 0.004 0.594 (0.67,0.99) (0.59,0.98) (0.62,0.97) (0.68,0.93) G|C|C|G|A 1.06 0.816 0.425 1.31g 0.105 0.969 1.23g 0.183 0.543 1.22h 0.142 0.746 (0.66,1.69) (0.95,1.81) (0.91,1.67) (0.94,1.60) G|C|C|G|G , , , 1.33 0.116 0.702 1.29 0.151 0.752 , , , (0.93,1.90) (0.91,1.81) G|T|C|G|A , , , 0.93 0.730 0.905 , , , , , , (0.63,1.39) G|T|C|G|G 1.41 0.007 0.248 0.63 0.028 0.821 0.86 0.690 0.028 1.01 0.984 0.014 (1.10,1.82) (0.41,0.95) (0.41,1.80) (0.61,1.66) G|T|T|A|A 1.01 0.947 0.731 1.19 0.573 0.029 1.07 0.590 0.062 1.06 0.579 0.118 (0.75,1.36) (0.66,2.14) (0.84,1.36) (0.87,1.29) aBlock 1 (rs475414|rs17299124), block 2 (rs693591|rs17300741|rs2078267), block 3. (rs3825018|rs475688|rs7932775|rs476037|rs478607) excluding frequency <0.01. bCombining the Eastern Polynesian (EP) low (EP(L)) and NZ European Caucasian sample sets. cCombining the EP high (EP(H)), Western Polynesian (WP) and EP/WP sample sets. dCombining the EP(H), EP(L), WP, and EP/WP sample sets. eCombining the EP(H), EP(L), WP, EP/WP and NZ European Caucasian sample sets. fOR: Inverse-variance weighted analysis odds ratio (adjusted for sex); P: inverse-variance weighted analysis P-value; Het P: heterogeneity P-value, inverse-variance weighted analyses were reanalyzed using a random-effects model if the Het P was <0.05. gEP/WP cohort was excluded from this meta-analysis owing to no female controls having the G|C|C|G|A haplotype.

lack of any known gene within the region (Ensembl release able to cause a difference in the regulation of SLC22A11 70, January 2013) suggests that the effect seen at or SLC22A12 expression. rs17299124 may operate by influencing the expression of a At block 2 (SLC22A11) rs2078267 has previously been nearby gene, perhaps SLC22A11. Previously rs2186571 has convincingly associated with gout in European Caucasian been associated with SU levels in the Pacific Micronesian (OR = 0.88, P =2.3×10-5; [2]) but the NZ European population of Kosrae [20]. This SNP, which maps several Caucasian sample set is inconsistent with these findings hundred kilobases upstream of SLC22A11, further impli- (OR = 0.98, P = 0.82). While low power may be a factor, cates this region in the etiology of gout. There are two in- we expected to observe an effect size consistent with that versions and a deletion spanning this region [29], possibly reported by Köttgen et al. [2] (OR = 1.14 to the C allele). http://arthritis-research.com/content/15/6/R220 Table 4 Sample set-specific haplotype association analysis at haplotype blocks 1 to 3 Flynn b c

East Polynesian (High) East Polynesian (Low) West Polynesian East/West Polynesian European Caucasian Therapy & Research Arthritis al. et a Hap Freq. ORd Pd Freq. ORd Pd Freq. ORd Pd Freq. ORd Pd Freq. ORd Pd (95% CI) (95% CI) (95% CI) (95% CI) (95% CI) Case Con Case Con Case Con Case Con Case Con Block 1 (rs475414|rs17299124) C|C 0.437 0.472 0.86 0.321 0.434 0.397 1.00 0.988 0.598 0.595 0.94 0.745 0.500 0.579 0.75 0.648 0.430 0.450 0.93 0.471 (0.63,1.16) (0.61,1.62) (0.67,1.34) (0.22,2.60) (0.76,1.14) T|A 0.024 0.051 0.35 0.017 0.179 0.203 0.81 0.525 0.004 0.044 0.12 0.009 0.026 0.053 0.33 0.461 0.262 0.245 1.10 0.393 (0.15,0.83) (0.42,1.56) (0.03,0.59) (0.02,6.19) (0.88,1.38) 1.34 0.063 0.387 0.400 1.18 0.562 0.396 0.361 1.24 0.228 0.474 0.368 1.53 0.486 0.302 0.301 1.00 0.984 2013, T|C 0.539 0.477 (0.98,1.82) (0.67,2.09) (0.88,1.76) (0.46,5.08) (0.80,1.24)

Block 2 (rs693591|rs17300741|rs2078267) 15 :R220 C|A|C 0.525 0.458 1.38 0.047 0.315 0.298 1.05 0.849 0.373 0.371 1.02 0.933 0.528 0.283 3.49 0.063 0.139 0.134 0.95 0.698 (1.00,1.90) (0.63,1.77) (0.72,1.43) (0.93,13.01) (0.72,1.25) G|A|C 0.299 0.350 0.77 0.136 0.287 0.272 1.13 0.654 0.436 0.461 0.83 0.288 0.278 0.478 0.38 0.106 0.329 0.302 1.07 0.543 (0.54,1.09) (0.66,1.95) (0.60,1.17) (0.12,1.23) (0.87,1.32) G|G|C 0.11 0.102 1.09 0.734 0.046 0.040 0.94 0.914 0.17 0.121 1.72 0.032 0.111 0.13 0.76 0.758 , , , , (0.66,1.83) (0.28,3.14) (1.05,2.83) (0.14,4.30) G|G|T 0.061 0.090 0.56 0.070 0.352 0.381 0.87 0.607 0.019 0.043 0.50 0.117 0.083 0.109 0.93 0.932 0.525 0.556 0.98 0.830 (0.30,1.05) (0.52,1.47) (0.21,1.19) (0.19,4.68) (0.81,1.18) Block 3 (rs3825018|rs475688|rs7932775|rs476037|rs478607) A|C|T|G|A 0.273 0.327 0.71 0.053 0.510 0.545 0.81 0.420 0.290 0.303 0.89 0.520 0.167 0.421 0.29 0.088 0.678 0.717 0.82 0.059 (0.50,1.01) (0.48,1.36) (0.61,1.28) (0.07,1.21) (0.67,1.01) G|C|C|G|A 0.179 0.136 1.26 0.283 0.078 0.115 0.78 0.582 0.159 0.129 1.37 0.214 0.139 0.053 , 0.997 0.035 0.024 1.19 0.534 (0.83,1.93) (0.32,1.88) (0.83,2.26) (0.68,2.08) G|C|C|G|G 0.116 0.108 1.14 0.599 0.039 0.038 0.82 0.767 0.142 0.114 1.54 0.100 0.111 0.105 1.71 0.659 , , , , (0.69,1.89) (0.23,2.98) (0.92,2.59) (0.16,18.73) G|T|C|G|A 0.057 0.068 0.84 0.574 0.000 0.016 , 0.998 0.111 0.102 1.01 0.971 0.139 0.053 0.96 0.968 , , , , (0.46,1.55) (0.59,1.73) (0.14,6.67) G|T|C|G|G 0.098 0.139 0.64 0.068 0.226 0.131 2.03 0.036 0.026 0.034 0.63 0.354 0.056 0.105 0.31 0.301 0.161 0.133 1.33 0.039 (0.40,1.03) (1.05,3.95) (0.24,1.68) (0.04,2.83) (1.01,1.75) G|T|T|A|A 0.266 0.213 1.53 0.030 0.128 0.131 0.90 0.776 0.264 0.314 0.78 0.183 0.361 0.263 2.28 0.300 0.088 0.087 1.04 0.836 ae8o 11 of 8 Page (1.04,2.24) (0.44,1.85) (0.55,1.12) (0.48,10.80) (0.74,1.44) aHap, haplotypes listed in alphabetical order; those with a frequency <0.01 were excluded. bEast Polynesian (High), individuals with a STRUCTURE ancestry level ≥0.67. cEast Polynesian (Low), individuals with a STRUCTURE ancestry level <0.67. dOR, odds ratio (adjusted by sex); P: asymptomatic P-value. freq., frequency; Con, control. Flynn et al. Arthritis Research & Therapy 2013, 15:R220 Page 9 of 11 http://arthritis-research.com/content/15/6/R220

Table 5 Genetic risk score analysis a a a b b Mean ± standard deviation ū1-ū2 t P OR P Variance explained (%)c Case Control (95% CI)a (95% CI) East Polynesian (H) 10.93 ± 1.56 10.45 ± 1.91 0.48 2.71 0.007 1.22 0.002 1.38 (0.13, 0.83) (1.07-1.39) East Polynesian (L) 4.81 ± 1.70 4.68 ± 1.96 0.13 0.47 0.64 1.03 0.77 0.09 (-0.44, 0.71) (0.84-1.26) West Polynesian 17.24 ± 1.21 16.57 ± 2.95 0.67 2.43 0.016 1.14 0.041 1.89 (0.13, 1.22) (1.01-1.30) Polynesian 11.16 ± 1.54 10.23 ± 2.04 0.93 1.54 0.14 1.32 0.21 5.11 (-0.31, 2.18) (0.85-2.05) European Caucasian 3.02 ± 1.60 2.94 ± 1.52 0.08 0.75 0.45 1.01 0.80 0.04 (-0.13, 0.28) (0.93-1.11) a b ū1-ū2, difference in means (case (ū1) minus control (ū2)); t, t-value, P, two-tailed t-test P-value. OR, odds ratio (adjusted by sex); P, logistic regression P-value. cBased on pseudo-R2 calculated during logistic regression.

It is possible that differences in phenotype among gout variant associated in the combined analysis (A-C-T-G-A; cases are important at SLC22A11. In the Köttgen et al. OR = 0.80, P = 0.004), with a consistent direction of effect study the association was restricted to the prevalent gout in Polynesians and European Caucasians. There was also a cases (OR = 0.86, P =4.4×10-6)andnotobservedinthe haplotype with ancestral specific effects; G-T-C-G-G was incident cases (OR = 0.96, P = 0.55). Ascertainment in the protective in the less admixed Polynesian sample sets, yet prevalent group was dominated by a combination of self- conferred risk in the more admixed EP and European report or use of urate-lowering medication and colchicine Caucasian sample sets. Our data therefore suggest that (prescribed for alleviating pain from acute gout), whereas multiple common variants within the SLC22A12 locus the incident group was ascertained, as were the NZ gout contribute to the risk of gout, in a population-dependent cases (OR = 0.98, P = 0.82), by use of the ARA classifi- manner. In the European Caucasian sample set there were cation criteria for gout, a superior ascertainment method nominal P-values for rs475688 and rs7932775 of 0.043 [30] (although the NZ cases were ascertained by clinical (OR = 1.26) and 0.033 (OR = 1.32), respectively. This is, to examination whereas the incident ones by a self-adminis- our knowledge, the first report of nominally significant as- tered questionnaire [31]). It is therefore counter-intuitive sociation of SLC22A12 with gout in European Caucasians. that the association of SLC22A11 should be restricted to However, given the likely weak effect, confirmation of this gout cases ascertained by less stringent criteria. Under- will require larger gout sample sets. standing this intriguing paradox may reveal new informa- The block-3 SNP rs475688, in moderate LD (r2 =0.80) tion on the role of SLC22A11 (OAT4) in the control of SU with rs3825018 in European Caucasians but low LD in and in the risk of gout. There are physiological data sho- CHB and Polynesians (r2 = 0.33 and 0.36, respectively) wing that OAT4 mediates hydrochlorothiazide-induced (Additional file 2: Figures S2 and S3), showed weak hyperuricemia [10] and also evidence for interaction of evidence for association with gout in the NZ European rs2078267 with diuretic use in determining the risk of Caucasian sample set (OR = 1.26, P = 0.04) consistent with gout [32]. It is possible that non-additive interaction with the minor (T) allele increasing SU levels in Caucasians diuretics could be a factor in the inconsistencies. In the (β =3.28 μmolL-1, P =1.3×10-17) [2]. This variant, Polynesian analysis there was nominal evidence for asso- however, has been associated with gout and SU in Han ciationwithgoutatrs2078267, with the minor allele Chinese and Solomon Island (Melanesian) sample sets with conferring risk (OR = 1.51, P = 0.022), consistent with the T allele conferring protection against gout (OR = 0.54, the urate-increasing effect of this allele in European P =1×10-4) and decreasing serum urate levels (β = −11.90 Caucasians [2]. However this association was not signifi- μmolL-1, P = 0.02) [15], a direction of association opposite cant when corrected for multiple testing. Given the lower to that observed in NZ European Caucasians. This oppos- allele frequency of this variant in Polynesians (<0.10 in the ite effect between Polynesians and European Caucasians less admixed samples), a larger sample size is required for at rs475688 is consistent with the opposing direction of more robust conclusions about the possible role of association seen with the block-3 G-T-C-G-G haplotype SLC22A11 in gout in Polynesian people. (rs475688 is the second marker) between less admixed At SLC22A12 the clearest findings came from analysis Polynesians, and European Caucasians and the more ad- of haplotype association. There was a gout-protective mixed Eastern Polynesians. It will be informative to test Flynn et al. Arthritis Research & Therapy 2013, 15:R220 Page 10 of 11 http://arthritis-research.com/content/15/6/R220

this haplotype for association with SU and gout in Asian Received: 5 July 2013 Accepted: 12 December 2013 and other Pacific populations. Published: 23 December 2013

Conclusions References We present several important findings: 1) novel asso- 1. Campion EW, Glynn RJ, DeLabry LO: Asymptomatic hyperuricemia. Risks ciation of rs17299124 upstream of SLC22A11 with gout in and consequences in the Normative Aging Study. Am J Med 1987, 82:421–426. Polynesians and not Caucasians, which is driven by a pro- 2. Köttgen A, Albrecht E, Teumer A, Vitart V, Krumsiek J, Hundertmark C, Pistis G, tective effect; 2) evidence for association of the SLC22A11 Ruggiero D, O’Seaghdha CM, Haller T, Yang Q, Tanaka T, Johnson AD, Kutalik Z, SNP rs2078267 with gout in Polynesians with the pre- Smith AV, Shi J, Struchalin M, Middelberg RP, Brown MJ, Gaffo AL, Pirastu N, Li G, Hayward C, Zemunik T, Huffman J, Yengo L, Zhao JH, Demirkan A, viously reported association in European Caucasians not Feitosa MF, Liu X, et al: Genome-wide association analyses identify 18 new replicated; and 3) the SLC22A12 analysis was the first loci associated with serum urate concentrations. Nat Genet 2013, report of nominally significant association with gout in 45:145–154. 3. Hollis-Moffatt JE, Xu X, Dalbeth N, Merriman ME, Topless R, Waddell C, European Caucasians, with the haplotype analysis sugges- Gow PJ, Harrison AA, Highton J, Jones PB, Stamp LK, Merriman TR: Role of ting multiple alleles conferring risk at SLC22A12 in an the urate transporter SLC2A9 gene in susceptibility to gout in New ancestry-specific manner. Zealand Maori, Pacific Island, and Caucasian case-control sample sets. Arthritis Rheum 2009, 60:3485–3492. 4. Matsuo H, Takada T, Ichida K, Nakamura T, Nakayama A, Ikebuchi Y, Ito K, Kusanagi Y, Chiba T, Tadokoro S, Takada Y, Oikawa Y, Inoue H, Suzuki K, Additional files Okada R, Nishiyama J, Domoto H, Watanabe S, Fujita M, Morimoto Y, Naito M, Nishio K, Hishida A, Wakai K, Asai Y, Niwa K, Kamakura K, Additional file 1: Table S1. Demographic and clinical characteristics of Nonoyama S, Sakurai Y, Hosoya T, et al: Common defects of ABCG2, a study participants. Table S2: Haplotype block summary. Table S3: high-capacity urate exporter, cause gout: a function-based genetic Association of block-1 to −3 single nucleotide polymorphisms (SNPs) with analysis in a Japanese population. Sci Transl Med 2009, 1:5ra11. serum urate in control individuals. Table S4: Interaction analysis between 5. Phipps-Green AJ H-MJ, Dalbeth N, Merriman ME, Topless R, Gow PJ, genotype, gout and co-morbid phenotypes. Harrison AA, Highton J, Jones PB, Stamp LK, Merriman TR: A strong role for Additional file 2: Figure S1. Intermarker linkage disequilibrium of the ABCG2 gene in susceptibility to gout in New Zealand Pacific Island single nucleotide polymorphisms (SNPs) previously associated with serum and Caucasian, but not Maori, case and control sample sets. Hum Mol – urate. Figure S2: Common linkage disequilbrium block haplotypes in Genet 2010, 19:4813 4819. European Caucasian and Han Chinese. Figure S3: Intermarker linkage 6. Hollis-Moffatt JE, Phipps-Green AJ, Chapman B, Jones GT, van Rij A, Gow PJ, disequilibrium between the 12 genoytped SNPs in European Caucasian Harrison AA, Highton J, Jones PB, Montgomery GW, Stamp LK, Dalbeth N, and Han Chinese. Figure S4: Intermarker linkage disequilibrium between Merriman TR: The renal urate transporter SLC17A1 locus: confirmation of the 12 genoytped SNPs in New Zealand sample sets. Figure S5: Power association with gout. Arthritis Res Ther 2012, 14:R92. calculations. 7. Urano W, Taniguchi A, Anzai N, Inoue E, Kanai Y, Yamanaka M, Endou H, Kamatani N, Yamanaka H: Sodium-dependent phosphate type 1 sequence polymorphisms in male patients with gout. Ann Rheum Dis 2010, 69:1232–1234. Abbreviations 8. Bleasby KCJ, Roberts CJ, Cheng C, Bailey WJ, Sina JF, Kulkarni AV, Hafey MJ, ARA: American Rheumatism Association; CEU: Centre d’Etude du Evers R, Johnson JM, Ulrich RG, Slatter JG: Expression profiles of 50 xenobiotic Polymorphisme Humain from Utah; CHB: Han Chinese from Beijing; transporter genes in humans and pre-clinical species: a resource for EP: Eastern Polynesian; GWAS: Genome-wide association studies; LD: Linkage investigations into drug disposition. Xenobiotica 2006, 36:963–988. disequilibrium; NZ: New Zealand; OAT4: Organic anion transporter 4; OR: Odds ratio; SNP: Single nucleotide polymorphism; SU: Serum urate; 9. Enomoto AKH, Chairoungdua A, Shigeta Y, Jutabha P, Cha SH, Hosoyamada M, URAT1: Urate transporter 1; WP: Western Polynesian; NZ: New Zealand. Takeda M, Sekine T, Igarashi T, Matsuo H, Kikuchi Y, Oda T, Ichida K, Hosoya T, Shimokata K, Niwa T, Kanai Y, Endou H: Molecular identification of a renal urate anion exchanger that regulates blood urate levels. Nature 2002, Competing interests 417:447–452. The authors declare that they have no competing interests. 10. Hagos YSD, Ugele B, Burckhardt G, Bahn A: Human renal organic anion transporter 4 operates as an asymmetric urate transporter. J Am Soc Nephrol 2007, 18:430–439. Authors’ contributions 11. Jang WCNY, Park SM, Ahn YC, Park SH, Choe JY, Shin IH, Kim SK: T6092C TJF, JEH-M, and TRM helped to design the study, oversee its execution, and polymorphism of SLC22A12 gene is associated with serum uric acid. prepare the manuscript. AP-G, MEM, RT, LKS and ND helped to provide Clin Chim Acta 2008, 398:140–144. clinical recruitment and prepare the manuscript. BC and GWM helped to 12. Jang WCNY, Ahn YC, Park SM, Yoon IK, Choe JY, Park SH, Her M, Kim SK: collect data and prepare the manuscript. All authors read and approved the G109T polymorphism of SLC22A12 gene is associated with serum uric final manuscript. acid level, but not with metabolic syndrome. Rheumatol Int 2012, 32:2257–2263. Acknowledgements 13. Kamatani Y, Matsuda K, Okada Y, Kubo M, Hosono N, Daigo Y, Nakamura Y, This work was supported by the Health Research Council of New Zealand, Kamatani N: Genome-wide association study of hematological and Arthritis New Zealand, New Zealand Lottery Health, Genetics Otago and the biochemical traits in a Japanese population. Nat Genet 2010, 4:210–215. University of Otago. The authors would like to thank Chris Franklin, Jill Drake, 14. Li C, Han L, Levin AM, Song H, Yan S, Wang Y, Wang Y, Meng D, Lv S, Ji Y, Roddi Laurence and Gabrielle Sexton for assistance in recruitment. Xu X, Liu X, Wang Y, Zhou L, Miao Z, Mi QS: Multiple single nucleotide polymorphisms in the human urate transporter 1 (hURAT1) gene are Author details associated with hyperuricaemia in Han Chinese. J Med Genet 2010, 1Department of Biochemistry, University of Otago, 364 Leith St, North 47:204–210. Dunedin 9016, New Zealand. 2Queensland Institute of Medical Research, 15. Tu HPCC, Lee CH, Tovosia S, Ko AM, Wang SJ, Ou TT, Lin GT, Chiang SL, Brisbane, Queensland, Australia. 3University of Otago, Christchurch, Chiang HC, Chen PH, Chang SJ, Lai HM, Ko YC: The SLC22A12 gene is New Zealand. 4Department of Medicine, University of Auckland, Auckland, associated with gout in Han Chinese and Solomon Islanders. Ann Rheum New Zealand. Dis 2010, 69:1252–1254. Flynn et al. Arthritis Research & Therapy 2013, 15:R220 Page 11 of 11 http://arthritis-research.com/content/15/6/R220

16. Stark KRW, Grassl M, Erdmann J, Schunkert H, Illig T, Hengstenberg C: 31. Choi HK, Atkinson K, Karlson EW, Willett W, Curhan G: Alcohol intake and Common polymorphisms influencing serum uric acid levels contribute risk of incident gout in men: a prospective study. Lancet 2004, to susceptibility to gout, but not to coronary artery disease. PLoS One 363:1277–1281. 2009, 4:e7729. 32. McAdams-DeMarco MA, Maynard JW, Baer AN, Kao LW, Kottgen A, Coresh J: 17. Guan M, Zhang J, Chen Y, Liu W, Kong N, Zou H: High-resolution melting A urate gene-by-diuretic interaction and gout risk in participants with analysis for the rapid detection of an intronic single nucleotide hypertension: results from the ARIC study. Ann Rheum Dis 2013, polymorphism in SLC22A12 in male patients with primary gout in China. 72:701–706. Scand J Rheumatol 2009, 38:276–281. 18. Tin AWO, Kao WH, Liu CT, Lu X, Nalls MA, Shriner D, Semmo M, Akylbekova EL, doi:10.1186/ar4417 Wyatt SB, Hwang SJ, Yang Q, Zonderman AB, Adeyemo AA, Palmer C, Meng Y, Cite this article as: Flynn et al.: Association analysis of the SLC22A11 Reilly M, Shlipak MG, Siscovick D, Evans MK, Rotimi CN, Flessner MF, Köttgen M, (organic anion transporter 4) and SLC22A12 (urate transporter 1) urate Cupples LA, Fox CS, Köttgen A, CARe and CHARGE Consortia: Genome-wide transporter locus with gout in New Zealand case-control sample sets Association Study for Serum Urate Concentrations and Gout among African reveals multiple ancestral-specific effects. Arthritis Research & Therapy Americans Identifies Genomic Risk Loci and a Novel URAT1 Loss-of-Function 2013 15:R220. Allele. Hum Mol Genet 2011, 20:4056–4068. 19. Taniguchi AUW, Yamanaka M, Yamanaka H, Hosoyamada M, Endou H, Kamatani N: A common mutation in an organic anion transporter gene, SLC22A12, is a suppressing factor for the development of gout. Arthritis Rheum 2005, 52:2576–2577. 20. Kenny EE, Kim M, Gusev A, Lowe JK, Salit J, Smith JG, Kovvali S, Kang HM, Newton-Cheh C, Daly MJ, Stoffel M, Altshuler DM, Friedman JM, Eskin E, Breslow JL, Pe’er I: Increased power of mixed models facilitates association mapping of 10 loci for metabolic traits in an isolated population. Hum Mol Genet 2011, 20:827–839. 21. Winnard D, Wright C, Taylor WJ, Jackson G, Te Karu L, Gow PJ, Arroll B, Thornley S, Gribben B, Dalbeth N: National prevalence of gout derived from administrative health data in Aotearoa New Zealand. Rheumatology (Oxford) 2012, 51:901–909. 22. Wallace SL, Robinson H, Masi AT, Decker JL, McCarty DJ, Yu TF: Preliminary criteria for the classification of the acute arthritis of primary gout. Arthritis Rheum 1977, 20:895–900. 23. 1000 genomes: a deep catalog of human genetic variation. [www.1000genomes.org] 24. Yang Q, Köttgen A, Dehghan A, Smith AV, Glazer NL, Chen MH, Chasman DI, Aspelund T, Eiriksdottir G, Harris TB, Launer L, Nalls M, Hernandez D, Arking DE, Boerwinkle E, Grove ML, Li M, Linda Kao WH, Chonchol M, Haritunians T, Li G, Lumley T, Psaty BM, Shlipak M, Hwang SJ, Larson MG, O’Donnell CJ, Upadhyay A, vanDuijnCM,HofmanA,et al: Multiple genetic loci influence serum urate levels and their relationship with gout and cardiovascular disease risk factors. Circ Cardiovasc Genet 2010, 3:523–530. 25. SNP function prediction. [http://snpinfo.niehs.nih.gov/snpinfo/snpfunc.htm] 26. Purcell S, Neale B, Todd, Brown K, Thomas L, Ferreira MA, Bender D, Maller J, Sklar P, de Bakker PI, Daly MJ, Sham PC: PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 2007, 81:559–575. 27. Stata | Data analysis and statistical software. [http://www.stata.com/] 28. Johnson GC, Esposito L, Barratt BJ, Smith AN, Heward J, Di Genova G, Ueda H, Cordell HJ, Eaves IA, Dudbridge F, Twells RC, Payne F, Hughes W, Nutland S, Stevens H, Carr P, Tuomilehto-Wolf E, Tuomilehto J, Gough SC, Clayton DG, Todd JA: Haplotype tagging for the identification of common disease genes. Nat Genet 2001, 29:233–237. 29. Korbel JO, Urban AE, Affourtit JP, Godwin B, Grubert F, Simons JF, Kim PM, Palejev D, Carriero NJ, Du L, Taillon BE, Chen Z, Tanzer A, Saunders AC, Chi J, Yang F, Carter NP, Hurles ME, Weissman SM, Harkins TT, Gerstein MB, Egholm M, Snyder M: Paired-end mapping reveals extensive structural variation in the human genome. Science 2007, 318:420–426. 30. Taylor W: Diagnosis of gout: considering clinical and research settings. Curr Rheum Rev 2011, 7:97–105. Submit your next manuscript to BioMed Central and take full advantage of:

• Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit